This is a 2-year grant application in response to PA-06-391: Research on Autism and Autism Spectrum Disorders R03. Clinical, animal, and genetic evidence points to a role for the hormone oxytocin in autism and suggests that oxytocin related abnormalities may be present in relatives of autism probands. Provocative clues raise a similar possibility for cholesterol. In the proposed pilot study, we plan to assess oxytocin and cholesterol levels in blood from a sample of 30 families (although 15 of these families will be financed through other means) each with a child diagnosed with autism or autism spectrum disorder and 20 families without autistic family members. The preliminary data obtained from this small project will help us design a larger one aimed ultimately at finding susceptibility genes for autism through the improved delineation of phenotypic expressions. Participating autism proband families will also have unaffected children. This will enable us to compare not only autism parents with control parents but also autism siblings with control siblings. We will screen all participants for autism and will subsequently use diagnostic assessments for potentially affected children. Furthermore, we will assess potential relationships between both oxytocin and cholesterol with behavioral and neuropsychological traits associated with the three core symptom domains of autism (social deficits, communication, and behavioral abnormalities), as well as executive functions. While many of these assessments have been used successfully in autism family studies;others have not been used in autism family studies before and are thus exploratory. Beyond their relationships with oxytocin and cholesterol, these measures will serve as a tool to evaluate more specific behavioral and neuropsychological traits, so called endophenotypes. Finally, using a new statistical method, multivariate distance matrix regression methodology, not previously employed in studies of autism or their relatives, we will try to identify subgroups within families, that is, families that can be described through a distinct endophenotypic profile. Clarifying endophenotypes and subgroups within autism families are both useful measures in the search for associated genes through molecular genetic studies. Accordingly, our study will contribute to the information necessary to define the causal factors of autism. PUBLIC HEALTH RELEVANCE: Because autism and autism spectrum disorder are genetically complex, the genes involved in this disorder are still largely unknown. We propose to conduct a study designed to isolate distinct behavioral and cognitive traits and their connections to oxytocin and cholesterol, two organic bio-chemicals that can be measured in the blood that are hypothesized to show abnormalities in autistic proband and their families. This will lead to very important new information that can inform and strengthen genetic studies of autism and ASD and thus help reveal the causes of this severe group of developmental disorders.